A time series approach will be taken to treated mental disorder with the only comprehensive case register in the United States, the Monroe County Register of New York State. The epidemiology of psychiatric admissions will be studied in a way which clarifies the contributions of trends, seasonal cycles, and community variables such as economic and deomographic change. Monthly admission counts will be generated for the 11 years from 1962 through 1972 for first and later admissions, public and private providers, in and outpatient treatment, and age, sex, and social status groups. In addition, suicide, economic, and other community monthly time series will be obtained. Time series analyses (including Box-Jenkins and other methods) will be used to address several questions regarding the relative contributions of predictors to admissions, the relationship of admissions and suicides, the relationship of different admissions series (e.g., public vs. private), the relationship of economic variables to admissions between demographic groups, and the relative importance of community predictors with and without detrending and deseasonalizing.